HistoART: Histopathology Artifact Detection and Reporting Tool

2025-06-23

AI Summary

Artifacts introduced during slide preparation and whole-slide imaging (WSI) acquisition severely compromise pathological analysis and diagnostic reliability. To address six common artifact types, this work systematically evaluates three detection approaches: (1) an end-to-end fine-tuned UNI foundation model, (2) a supervised ResNet50 classifier, and (3) a traditional method integrating texture, color, and frequency-domain features. Notably, the UNI-based approach—first applied to WSI artifact detection—leverages a general-purpose foundation model, markedly enhancing generalizability and robustness. Validated on 50,000 WSIs, it achieves an AUROC of 0.995, significantly outperforming baselines. Complementing detection, we introduce an interpretable quality scoring card that visualizes artifact type, spatial location, and severity. This work establishes a new paradigm for WSI quality control: highly accurate, inherently interpretable, and production-ready.

Technology Category

• Computer Vision: Other Foundations of Computer Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

In modern cancer diagnostics, Whole Slide Imaging (WSI) is widely used to digitize tissue specimens for detailed, high-resolution examination; however, other diagnostic approaches, such as liquid biopsy and molecular testing, are also utilized based on the cancer type and clinical context. While WSI has revolutionized digital histopathology by enabling automated, precise analysis, it remains vulnerable to artifacts introduced during slide preparation and scanning. These artifacts can compromise downstream image analysis. To address this challenge, we propose and compare three robust artifact detection approaches for WSIs: (1) a foundation model-based approach (FMA) using a fine-tuned Unified Neural Image (UNI) architecture, (2) a deep learning approach (DLA) built on a ResNet50 backbone, and (3) a knowledge-based approach (KBA) leveraging handcrafted features from texture, color, and frequency-based metrics. The methods target six common artifact types: tissue folds, out-of-focus regions, air bubbles, tissue damage, marker traces, and blood contamination. Evaluations were conducted on 50,000+ image patches from diverse scanners (Hamamatsu, Philips, Leica Aperio AT2) across multiple sites. The FMA achieved the highest patch-wise AUROC of 0.995 (95% CI [0.994, 0.995]), outperforming the ResNet50-based method (AUROC: 0.977, 95% CI [0.977, 0.978]) and the KBA (AUROC: 0.940, 95% CI [0.933, 0.946]). To translate detection into actionable insights, we developed a quality report scorecard that quantifies high-quality patches and visualizes artifact distributions.

Problem

• Detects artifacts in Whole Slide Imaging for cancer diagnostics
• Compares three artifact detection methods for histopathology slides
• Evaluates performance on 50,000+ image patches from diverse scanners

Innovation

• Fine-tuned UNI foundation model for artifact detection
• ResNet50-based deep learning for WSI analysis
• Knowledge-based metrics for texture and color features